AI agent personality switching method and device based on NFC identification

Abstract

The invention relates to the technical field of artificial intelligence and Internet of Things equipment, and discloses an AI (artificial intelligence) agent personality switching method based on NFC (near field communication) identification, which realizes dynamic switching of AI agent personality in a physical mode, and comprises the following steps: S1, an equipment end obtains and decrypts an NFC label signal of an agent; s2, the equipment end obtains an AI virtual role ID corresponding to the intelligent agent according to the decryption information, and sends a request parameter to a cloud server; s3, the cloud server obtains personality configuration of the intelligent agent from a role database of the cloud server according to the request parameters and returns the personality configuration to the equipment end, wherein the personality configuration comprises an image, voice tone, speed, tone, role background, story index information and a knowledge template; and S4, the device side updates and switches the personality state corresponding to the intelligent agent according to the personality configuration, and performs voice interaction with the user according to the current personality state parameter. The invention further discloses an AI agent personality switching device for implementing the AI agent personality switching method.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence and Internet of Things (IoT) devices, and in particular to a method and apparatus for switching the personality of an AI intelligent agent based on NFC recognition. Background Technology

[0002] Currently, various smart voice devices exist on the market, such as story machines, smart speakers, and interactive doll devices. These devices mostly use a unified AI voice model to achieve voice interaction with users or play fixed story content. However, these smart voice devices typically have a single, fixed AI voice model or personality built in. This means that when interacting with users, regardless of changes in the scene or the interaction object, they use a uniform voice style, knowledge base, and interaction logic, failing to provide differentiated role-playing experiences based on needs and making it difficult to meet users' demands for diverse and personalized role-playing. Some devices simulate role switching by playing different local story files, but this is merely a content switch, not a switch in the AI ​​personality itself. This type of role switching makes it difficult to achieve real-time mapping and dynamic updates between the cloud server personality and the physical character, failing to meet the personalized and differentiated design requirements at the AI ​​agent level, resulting in a subpar user experience. Summary of the Invention

[0003] Therefore, it is necessary to address the above-mentioned shortcomings by providing an NFC-based method and device for dynamically switching the personality of an AI agent through physical means.

[0004] A method for switching the personality of an AI intelligent agent based on NFC recognition includes the following steps: S1. The device acquires and decrypts the NFC tag signal of the smart agent; S2. The device obtains the AI ​​virtual role ID corresponding to the intelligent agent based on the decryption information and sends request parameters to the cloud server. S3. The cloud server retrieves the personality configuration of the intelligent agent from its role database according to the request parameters and returns it to the device. The personality configuration includes appearance, voice tone, speech rate, timbre, role background, story index information, and knowledge template. S4. The device updates and switches the personality state corresponding to the intelligent agent according to the personality configuration, and interacts with the user via voice according to the current personality state parameters.

[0005] In one embodiment, the device acquiring and decrypting the smart agent's NFC tag signal includes: S11. The device detects the NFC tag signal of the smart agent, confirms the existence of the smart agent's NFC tag, and sends a signal for the next decryption action. S12. The device receives the decryption action signal, reads the encrypted area information specified by the smart agent's NFC tag according to the key information, and uses the key to decrypt the read encrypted information. The device obtains the AI ​​virtual character ID corresponding to the intelligent agent based on the decrypted information, and sends a request to the cloud server including: S21. The device obtains the successfully decrypted plaintext information, extracts information of a specific length from a specified position, and obtains the AI ​​virtual role ID corresponding to the intelligent agent. S22. The device re-encrypts the AI ​​virtual character ID and packages it into request parameters, and then accesses the AI ​​interface of the third-party cloud server. The cloud server retrieves the agent's personality configuration from the role database based on the request parameters and returns it to the device, including: S31. The cloud server obtains and parses the request parameters sent by the device. S32. Determine whether the AI ​​virtual character ID in the request parameters is consistent with the AI ​​virtual character ID currently used on the device. If they are consistent, maintain the current personality configuration and send a role existence notification to the device. Otherwise, proceed to step S33. S33. Search the role database. If the corresponding AI virtual role ID is found, return the personality configuration corresponding to the AI ​​virtual role ID to the device and send a role existence notification to the device. Otherwise, send a no-role-existence notification to the device. In step S4, while the device interacts with the user via voice based on the current personality state parameters, it also displays animations and content corresponding to the personality state and voice.

[0006] In one embodiment, prior to step S1, custom role creation is also included: Configure role parameters on the cloud server and send the role parameters to the devices that interact with the cloud server; The device writes the role parameters into the smart agent's NFC tag in the form of an AI virtual role ID.

[0007] In one embodiment, step S4 is followed by a background switch for the agent's image: Capture images containing environmental scenes and / or clothing, and convert the content of the images into digital information; The device receives the digital information and writes the image material information from the digital information into the image data area of ​​the smart agent's NFC tag; The device re-reads the AI ​​virtual role ID of the intelligent agent and reports the intelligent agent information to the cloud server; The cloud server acquires intelligent agent information, parses and applies new scene data and / or image data from the intelligent agent's image data area; The cloud server switches the scene data and / or appearance data of the intelligent agent, sends the new scene and / or appearance of the role to the device, and the device displays the new scene and / or appearance of the intelligent agent.

[0008] In one embodiment, step S4 is followed by an agent role growth update: The device acquires location information and writes it into the location information area of ​​the smart agent's NFC tag; The device re-reads the AI ​​virtual role ID of the intelligent agent and reports the intelligent agent information to the cloud server; The cloud server acquires and parses the location information content in the location information area of ​​the intelligent agent, determines whether there is any new location information, and updates the agent's location information when there is new location information; The cloud server provides the new location information of the character to the AI ​​character model as setting information. The AI ​​character model learns and updates its growth information based on the new location information to simulate an intelligent agent model that travels and grows with the user. The growth information includes the local customs and culture of the area where the new location is located.

[0009] In one embodiment, the device enters a sleep state for a preset duration when it does not detect an NFC tag signal, fails to resolve the NFC tag, fails to recognize the NFC tag, or fails to recognize the NFC tag successfully less than a preset value.

[0010] In one embodiment, the device caches the most recent personality configuration while offline.

[0011] The present invention also discloses an AI intelligent agent personality switching device based on NFC identification, which is used to implement the above-mentioned AI intelligent agent personality switching method based on NFC identification. The AI ​​intelligent agent personality switching device includes a device end that interacts with the intelligent agent and a cloud server that interacts with the device. The device includes: The NFC read / write module is used to read the NFC tag signal of the smart agent and to perform write operations to the smart agent to update the NFC tag signal of the smart agent. The information processing module is used to parse the NFC tag signal of the intelligent agent and obtain the AI ​​virtual role ID corresponding to the intelligent agent; The communication module is used to send request parameters containing the AI ​​virtual character ID to the cloud server and receive the personality configuration returned by the cloud server to switch and update the corresponding personality state of the intelligent agent. The audio module is used to acquire and process user voice and play AI voice messages. The cloud server has a character database for storing character personality configurations. The cloud server retrieves the personality configuration of the agent from its character database according to the request parameters and returns it to the communication module.

[0012] In one embodiment, the device further includes: The display module is used to display the AI ​​character's appearance, interactive animations, and content. The interaction module provides an interface for interacting with users.

[0013] In one embodiment, the communication module interacts with the cloud server via an SDK module. The NFC-based AI agent personality switching method and apparatus of this invention associates a device with agents equipped with different NFC tags. By having the device touch the NFC tags of different agents, the device dynamically loads and activates the corresponding personalized AI personality on the cloud server, obtaining a personality configuration corresponding to the agent's NFC tag signal. This allows a single device to provide multiple immersive character interaction experiences. The personality configuration sets the character's image, voice tone, speech rate, timbre, background, story index information, and knowledge templates, ensuring that each character not only has a different voice, but also that their language style, knowledge background, dialogue logic, and emotional expression highly match their own background story, effectively meeting users' diverse needs for characters and enhancing the user experience. Attached Figure Description

[0014] Figure 1 This is a flowchart of an AI agent personality switching method in one embodiment of the present invention; Figure 2 This is a flowchart illustrating the interaction between local NFC identification and the cloud server in one embodiment of the present invention; Figure 3 This is a flowchart illustrating the AI ​​personality switching logic of a cloud server in one embodiment of the present invention. Figure 4 A flowchart illustrating the creation of a custom role in one embodiment of the present invention; Figure 5 This is a flowchart illustrating the background switching process for an intelligent agent in one embodiment of the present invention. Figure 6 This is a flowchart illustrating the growth and update process of an agent role in one embodiment of the present invention; Figure 7 This is a module connection diagram of an AI agent personality switching device in one embodiment of the present invention; Figure 8 This is a block diagram of instruction execution for an AI agent personality switching device in one embodiment of the present invention. Detailed Implementation

[0015] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0016] Example 1 Please see Figure 1 This invention discloses an NFC-based method for dynamically switching the personality of an AI agent through physical means. This method is applicable to interactions between the AI ​​agent and devices such as children's interactive story machines, voice assistants, and interactive doll devices. The AI ​​agent can be a doll or other physical toy with a built-in NFC tag and a specific image. Each AI agent's NFC tag stores its unique identifier ID, which is the corresponding AI virtual character ID. Preferably, the device is a story machine. Specifically, the AI ​​agent personality switching method includes the following steps: S1. The device acquires and decrypts the NFC tag signal of the smart agent.

[0017] Furthermore, the device acquires and decrypts the smart agent's NFC tag signal, including: S11. The device detects the NFC tag signal of the smart agent, confirms the existence of the smart agent's NFC tag, and sends a signal for the next decryption action.

[0018] S12. The device receives the decryption action signal, reads the encrypted area information specified by the smart agent's NFC tag according to the key information, and uses the key to decrypt the read encrypted information.

[0019] In this embodiment, the device obtains the NFC tag signal of the smart agent through NFC signal identification and decrypts the NFC tag signal. In this way, the NFC signal is transmitted in encrypted form, which can prevent the smart agent's NFC tag signal from being sniffed and copied by other devices, prevent the smart agent's NFC tag from being illegally copied, and the automatic verification of the decryption process can ensure that the data is not modified during transmission, thus ensuring the reliability of signal transmission.

[0020] S2. The device obtains the AI ​​virtual role ID corresponding to the intelligent agent based on the decrypted information and sends request parameters to the cloud server. In other words, the request parameters sent by the device to the cloud server include, on the one hand, a request to obtain the role's personality configuration, and on the other hand, the AI ​​virtual role ID of the intelligent agent, so as to obtain the personality configuration of the role associated with the AI ​​virtual role ID from the cloud server.

[0021] The device obtains the AI ​​virtual role ID corresponding to the intelligent agent based on the decrypted information and sends a request to the cloud server including: S21. The device obtains the successfully decrypted plaintext information, extracts a specific length of information content at a specified position, and obtains the AI ​​virtual role ID corresponding to the intelligent agent.

[0022] S22. The device re-encrypts the AI ​​virtual character ID, packages it into request parameters, and accesses the AI ​​interface of the third-party cloud server.

[0023] In other words, the process of the device requesting data from the cloud server is also encrypted. This ensures the data is encrypted at the device level, preventing tampering or loss during transmission. Furthermore, the device encrypts, encapsulates, and verifies the data first, while the cloud server only handles decryption, storage, and comparison. The cloud server does not bear all the encryption calculations, reducing its computational burden and improving system security and scalability. Simultaneously, the cloud server can quickly verify whether the data originated from a legitimate device and whether it is forged, achieving dual trusted authentication of both device identity and data content.

[0024] In this embodiment, when the device does not detect an NFC tag signal, fails to resolve the NFC tag, has inconsistent NFC tag identification, or has fewer than a preset number of successful NFC tag identifications, it enters a sleep state for a preset duration. After the sleep state ends, the device can trigger the NCF identification process again after contacting the smart agent. This reduces the number of invalid retries, lowers device power consumption, and since a single identification error is often a momentary interference, retrying after sleep avoids the interference window, reducing continuous misjudgments and repeated triggers, making identification more stable and reliable, improving identification stability and system security, and reducing the computational and communication pressure on the device and cloud server.

[0025] For details, please refer to Figure 2 During the process of local NFC identification and interaction with the cloud server (i.e., interaction between the device and the cloud server), when the smart agent (i.e. Figure 2When the NFC tag doll shown is brought near the device, the device's NFC read / write module is triggered. The NFC read / write module checks for the presence of an NFC tag signal. If not, the device sleeps for 100ms and then checks again. If an NFC tag signal is present, the NFC read / write module performs key verification with the NFC tag. If key verification fails, the NFC recognition success count is decremented by 1, and the NFC tag signal detection is repeated after a 50ms sleep. If key verification succeeds, the device reads the encrypted area information from the smart agent's NFC tag and decrypts it using the built-in key. If decryption fails, the NFC recognition success count is decremented by 1, and the NFC tag signal detection is repeated after a 50ms sleep. If decryption succeeds, the device extracts the valid information from the NFC tag and compares the acquired current NFC tag information with the NFC tag information stored locally on the device. If the current NFC tag information does not match the NFC tag information stored locally on the device, the locally stored NFC tag information is replaced with the acquired current NFC tag information, the NFC tag recognition success count is reset to 1, and the NFC tag signal detection is repeated after a 50ms sleep. If the current NFC tag information matches the NFC tag information stored locally on the device, the locally stored NFC tag information is replaced with the acquired current NFC tag information, and the NFC tag recognition success count is incremented by 1. Then, it is determined whether the NFC tag recognition success count is less than a preset value (e.g., 3 times). If the NFC tag recognition success count is less than 3 times, the device sleeps for 50ms and then re-detects the NFC tag signal. If the NFC tag recognition success count is greater than or equal to 3 times, the AI ​​agent ID is identified, and specific position and length information from the NFC tag is extracted to obtain the AI ​​virtual role ID. Next, the AI ​​virtual role ID is packaged and encrypted by the device's information processing module, and the encrypted information packet is uploaded to the cloud server (or cloud server) via the device's communication module.

[0026] S3. The cloud server retrieves the agent's personality configuration from its role database based on the request parameters and returns it to the device. The personality configuration includes appearance, voice tone, speech rate, timbre, role background, story index information, and knowledge template.

[0027] It should be noted that in this embodiment, the role database of the cloud server stores the personality configurations of multiple roles. Each role has a different personality configuration, and different roles are associated with different AI virtual role IDs. In this way, the cloud server can call the corresponding personality configuration according to the AI ​​virtual role ID signal in the request parameters, so that the personality state on the device can be adapted to the matching intelligent agent, thereby realizing the personality switching of the intelligent agent.

[0028] Please see Figure 3The cloud server retrieves the agent's personality configuration from the role database based on the request parameters and returns it to the device. This configuration includes: S31. The cloud server obtains and parses the request parameters sent by the device.

[0029] S32. Determine whether the AI ​​virtual character ID in the request parameters is consistent with the AI ​​virtual character ID currently used on the device. If they are consistent, maintain the current personality configuration and send a role existence notification to the device. Otherwise, proceed to step S33.

[0030] S33. Search the character database. If the corresponding AI virtual character ID is found, return the personality configuration corresponding to the AI ​​virtual character ID to the device and send a character existence notification to the device. Otherwise, send a no-character-existence notification to the device.

[0031] Further integration Figure 2 and Figure 3 After receiving the request parameters in the information packet, the cloud server parses the information packet (i.e., the data packet), obtains the AI ​​virtual character ID of the role, and retrieves the character ID from the character database. If the corresponding character ID is not found in the character database, the cloud server sends a notification that the AI ​​character does not exist to the device. If the corresponding character ID is found in the character database, the cloud server configures the corresponding character parameters, sets parameters such as the intelligent agent character's appearance, voice tone, speech rate, timbre, character background, story index information, and knowledge template, and sends a notification that the character exists to the device. In this embodiment, when a new role ID is matched with a role ID on the role server, it is first determined whether the new role ID is consistent with the current role ID on the role server. That is, it is determined whether the AI ​​virtual role ID of the agent is consistent with the AI ​​virtual role ID currently being executed on the device. If they are consistent, the current personality state is maintained. If they are inconsistent, a matching role ID is searched in the role server, and the AI ​​virtual role ID on the device is switched and updated. If there is no matching role ID in the role server, it means that the current agent and the device are not associated. In this case, a new role ID can be created in the cloud server to enable the current agent to match and interact with the device.

[0032] S4. The device updates and switches the personality state corresponding to the intelligent agent based on the personality configuration, and interacts with the user via voice based on the current personality state parameters. Specifically, after obtaining the personality configuration returned by the cloud server, the device (i.e., the main device) provides a voice prompt indicating that there is no AI role, or engages in AI dialogue interaction with the user via voice. In this way, by associating a single physical device (the device) with intelligent agents of different shapes (such as dolls or cards) embedded with NFC tags, the user can dynamically load and activate the personalized AI personality corresponding to the cloud server (e.g., Sun Wukong, scientist, prince, princess, etc.) by touching the intelligent agent with different NFC tags. In addition, in this solution, the identifier ID of a specific AI role (i.e., AI virtual role ID) is pre-embedded in the NFC tag. Utilizing its near-field communication characteristics, one-click role switching is achieved. The user only needs to touch the NFC tag carrying the different roles with the device to directly, seamlessly, and accurately complete the switching of the AI ​​personality, simplifying complex software or menu operations into a one-step natural physical interaction. The parameters of the AI ​​characters are stored on a cloud server, and each AI character is assigned a unique identifier ID (i.e., AI virtual character ID). The device only acts as an interaction medium, binding the AI ​​character's identifier ID to the cloud-based AI personality. This allows a simple physical operation (such as touching NFC) to load and activate a specific AI personality corresponding to a physical object (intelligent agent) on the cloud server, achieving real-time and dynamic binding between the physical world and the cloud-based digital personality. In step S4, while the device interacts with the user via voice based on the current personality state parameters, it also displays animations and content corresponding to the personality state and voice, thus showcasing the AI ​​character's image and enhancing the user experience.

[0033] Please see Figure 4 Before step S1, the process also includes custom character creation: The process involves configuring role parameters on a cloud server and then sending these parameters to the device interacting with the cloud server. The device then writes these parameters into the smart agent's NFC tag as an AI virtual character ID. Specifically, the cloud server first sets a custom character background and parameters, including voice, personality, gender, and age, and then sets the character's appearance, thus creating the cloud-based character. Next, a unique character ID is generated for the created cloud-based custom character's parameters. The character data is then encrypted and packaged, and the encrypted and packaged data packet is sent to the device interacting with the cloud server. After receiving the character data, the device encrypts it with a private key and then uses its NFC read / write module to write the character data into the smart agent's NFC tag, completing the custom character creation. In this way, while updating the smart agent's NFC tag, the cloud server also stores the corresponding character ID, allowing the smart agent to trigger the appropriate personality when it interacts with the device. In this embodiment, a complete closed loop from digital creation to physical carrier is constructed through the NFC read / write module on the device. Users can freely create their own AI personality on the cloud server platform, setting its voice, personality, and appearance. After creation, the system generates a corresponding character data package, which can be bound to a blank or user-specified NFC tag through the NFC read / write module on the device. This permanently "injects" a user-defined digital character into the physical intelligent agent, achieving true personalization. The definition, learning, and updating of the AI ​​personality are all handled on the cloud server. The device uses the character's unique identifier ID to bind with the cloud-based AI personality to add new characters. The optimization of existing characters can also be completed in real time in the cloud without replacing or upgrading hardware. Users can unlock new personalities without purchasing new physical dolls, reducing the cost of using the intelligent agent.

[0034] Please see Figure 5 Step S4 also includes switching the background of the agent's image: Capture images containing environmental scenes and / or clothing, and convert the content of the images into digital information; The device receives digital information and writes the image material information from the digital information into the image data area of ​​the smart agent's NFC tag; The device re-reads the AI ​​virtual role ID of the intelligent agent and reports the intelligent agent information to the cloud server; The cloud server acquires intelligent agent information, parses and applies new scene data and / or image data from the intelligent agent's image data area; The cloud server switches the scene data and / or appearance data of the intelligent agent, sends the new scene and / or appearance of the role to the device, and the device displays the new scene and / or appearance of the intelligent agent.

[0035] In this embodiment, the device can receive image materials acquired by an external camera device, or the device itself can have a built-in camera module, such as a mobile phone with a camera function. Specifically, image materials are captured within a mobile app associated with the device, including clothing, environmental scenes, hairstyles, etc. The image materials are converted into digital information and sent to the device (or the device's camera module can capture images and convert them into digital information, which is then sent to the device's information processing module). After receiving this information, the device uses a data private key to encrypt the digital information of the image and calls the NFC read / write module to overwrite the digital information data onto the image data area of ​​the smart agent's NFC tag. After the data is successfully written to the NFC tag, the device reads the NFC tag information from the smart agent again and then uploads the NFC tag information to the cloud server. The cloud server parses the NFC tag information, obtains the character's image data area data, and generates a new image and scene based on the image data, thereby switching the AI ​​character's new image and scene. After the switch, the cloud server sends the new image and scene information to the device, and the device's display module displays the new character image and scene. This expands the dimensions of character customization, allowing users to personalize the appearance attributes of AI characters (such as clothing, accessories, and scene backgrounds). Users can photograph real-world clothing or scenes using mobile devices, and after image recognition and digitization processing by a dedicated app, a corresponding image data package is generated. Finally, this image data package is updated to the smart agent's NFC tag via the device's NFC writing function. When the device interacts with the cloud server, it can load this customized image, presenting a unique visual experience on the screen or in the accompanying app, in addition to voice interaction, thus achieving deep personalization through "audiovisual integration" to adapt to changes in user aesthetics or the needs of role-playing.

[0036] Please see Figure 6 Step S4 also includes the growth and update of the agent's role: The device acquires location information and writes it into the location information area of ​​the smart agent's NFC tag; The device re-reads the AI ​​virtual role ID of the intelligent agent and reports the intelligent agent information to the cloud server; The cloud server acquires and parses the location information content in the location information area of ​​the intelligent agent, determines whether there is any new location information, and updates the agent's location information when there is new location information; The cloud server provides the new location information of the character to the AI ​​character model as setting information. The AI ​​character model learns and updates its growth information based on the new location information to simulate an intelligent agent model that travels and grows together with the user. The growth information includes the local customs and culture of the area where the new location is located.

[0037] Specifically, the mobile app associated with the device acquires geolocation information. This geolocation information can be the real-time geolocation information of the user and the smart agent, or it can be geolocation information randomly entered by the user. The acquired geolocation information is packaged, encrypted, and sent to the device. The device uses a data private key to encrypt the location information and drives the NFC read / write module to write the encrypted location information to the location information area of ​​the smart agent's NFC tag. After successful writing, the device rereads the smart agent's NFC tag information and uploads it to the cloud server. The cloud server parses the smart agent's NFC tag information and obtains the smart agent's geolocation data. It compares this geolocation data with previously saved geolocation information. If the geolocation information remains unchanged, the current personality configuration is maintained. If the geolocation information changes, it is updated. This change includes two scenarios: the new geolocation information is not the current geolocation information but is stored on the cloud server, or the new geolocation information is not the current geolocation information and is not stored on the cloud server. After updating the geolocation information, the geolocation is provided as configuration information to the intelligent agent model on the cloud server. This allows the cloud server to enrich the AI's dialogue content, background knowledge, and worldview based on the local customs and culture corresponding to the geolocation in the intelligent agent model. This simulates a digital life form that shares a life journey with the user and grows together, thus completing the AI ​​character's development. In this way, through the NFC read / write function on the device, the physical intelligent agent is transformed into a "memory carrier" recording the user's real-life trajectory. The device automatically and continuously writes the user's visited geolocation information (such as countries and cities) as important growth data into the intelligent agent's NFC tag. During each interaction, the cloud-based AI character reads this accumulated location data and enriches its dialogue content, background knowledge, and worldview accordingly, demonstrating its knowledge and experiences of "having been to" these places. This makes the AI ​​character no longer a fictional scripted character, but a digital life form that shares a life journey with the user and grows together.

[0038] This solution can be understood as follows: It not only utilizes the device's NFC read / write module to read and write the unique ID of the locally encrypted smart agent NFC tag, parsing and writing the corresponding AI virtual character ID, allowing the device to automatically call the cloud API using the AI ​​virtual character ID to obtain the corresponding character's AI personality configuration; it also allows for writing operations on the smart agent. Through the device's NFC read / write module's writing function, users can replace the original character ID with a custom new one, thus changing the doll's character. Furthermore, in this embodiment, users can choose to add a unique identifier written to the device's name after the original character ID, enabling the same doll to display different images on different devices, fulfilling the device's personalized needs for the smart agent's image. For some smart agent images and AI personalities that are entirely user-owned, users can configure the smart agent image and AI character personality on the cloud server platform, then write the smart agent character ID and device ID into the smart agent's NFC tag, binding the smart agent to a single device and ensuring that the image and AI character personality corresponding to that smart agent can only be accessed by the user's own device. This allows for the binding and switching of AI agent personality and physical role, meets users' personalized requirements, and ensures data security due to the added encrypted writing function.

[0039] In addition, the device caches the most recent personality configuration when offline. This way, when users use the device offline, the AI ​​character with the most recently cached personality configuration can interact with the user, thus maintaining the user's interactive experience.

[0040] Example 2 Please combine Figure 1-8This invention also discloses an AI agent personality switching device based on NFC recognition, which is used to implement the above-mentioned AI agent personality switching method based on NFC recognition. The AI ​​agent personality switching device includes a device terminal 20 that interacts with the agent 10 and a cloud server 30 that interacts with the device. The device terminal 20 includes an NFC read / write module 201, an information processing module 202, a communication module 203, and an audio module 204. The NFC read / write module 201 is used to read the NFC tag signal of the agent 10 and to perform a write operation to the agent 10 to update the NFC tag signal of the agent 10. The information processing module 202 is used to parse the NFC tag signal of the agent 10 and obtain the AI ​​virtual role ID corresponding to the agent 10. The communication module 203 is used to send a request parameter containing the AI ​​virtual role ID to the cloud server 30 and receive the personality configuration returned by the cloud server 30 to switch and update the personality state corresponding to the agent 10. The audio module 204 is used to acquire and process the user's voice and play AI voice. In this embodiment, the audio module 204 includes an audio acquisition module 2041, a voice processing module 2042, and a voice playback module 2043. The audio acquisition module 2041 acquires and records user voice; the voice processing module 2042 processes the user's interaction with the AI ​​voice based on the personality configuration; and the voice playback module 2043 plays the AI ​​voice based on the processing result of the voice processing module 2042. The cloud server 30 has a role database for storing personality configurations. The cloud server 30 retrieves the personality configuration of the intelligent agent 10 from its role database according to request parameters and returns it to the communication module 203. In this embodiment, the cloud server 30 stores AI personality, voice tone, story background, and other content. The communication module 203 uploads or downloads data from the cloud. Preferably, the communication module 203 interacts with the cloud server 30 through an SDK module. The information processing module 202 parses and packages the data. The audio acquisition module 2041 is a microphone recording module, and the voice playback module 2043 is a speaker module.

[0041] Furthermore, the device 20 also includes a display module 205 and an interaction module 206. The display module 205 is used to display the AI ​​character's image, interactive animations, and content, and can be an LCD screen. The interaction module 206 provides an interface for interaction between the device 20 and the user, allowing the user to send commands to the device 20 or interact with the AI ​​character displayed on the device 20. Additionally, in this embodiment, an NFC read / write module 201 is located in the base area of ​​the device 20 and is used to identify the NFC tag of the placed intelligent agent 10.

[0042] It should be noted that the working process of the AI ​​agent personality switching device in this embodiment is exactly the same as the process of the AI ​​agent personality switching method in Embodiment 1. For details, please refer to the relevant description in Embodiment 1, which will not be repeated here.

[0043] The AI ​​intelligent agent personality switching method and device based on NFC identification of the present invention associates device 20 with intelligent agents 10 equipped with different NFC tags. By having device 20 touch the NFC tags of different intelligent agents 10, device 20 can dynamically load and activate the personalized AI personality corresponding to cloud server 30, and obtain the personality configuration corresponding to the NFC tag signal of intelligent agent 10. This enables one device 20 to provide multiple immersive role interaction experiences. Through personality configuration, the image, voice tone, speech speed, timbre, character background, story index information, and knowledge template of the character are set, so that each character not only has a different voice, but its language style, knowledge background, dialogue logic and emotional expression are highly consistent with the character's own background story, effectively meeting the user's diverse needs for characters and improving the user experience.

[0044] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0045] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for switching the personality of an AI intelligent agent based on NFC recognition, characterized in that, Includes the following steps: S1. The device acquires and decrypts the NFC tag signal of the smart agent; S2. The device obtains the AI ​​virtual role ID corresponding to the intelligent agent based on the decryption information and sends request parameters to the cloud server. S3. The cloud server retrieves the personality configuration of the intelligent agent from its role database according to the request parameters and returns it to the device. The personality configuration includes appearance, voice tone, speech rate, timbre, role background, story index information, and knowledge template. S4. The device updates and switches the personality state corresponding to the intelligent agent according to the personality configuration, and interacts with the user via voice according to the current personality state parameters.

2. The AI ​​intelligent agent personality switching method based on NFC recognition according to claim 1, characterized in that, The device acquires and decrypts the NFC tag signal of the smart agent, including: S11. The device detects the NFC tag signal of the smart agent, confirms the existence of the smart agent's NFC tag, and sends a signal for the next decryption action. S12. The device receives the decryption action signal, reads the encrypted area information specified by the smart agent's NFC tag according to the key information, and uses the key to decrypt the read encrypted information. The device obtains the AI ​​virtual character ID corresponding to the intelligent agent based on the decrypted information, and sends a request to the cloud server including: S21. The device obtains the successfully decrypted plaintext information, extracts information of a specific length from a specified position, and obtains the AI ​​virtual role ID corresponding to the intelligent agent. S22. The device re-encrypts the AI ​​virtual character ID and packages it into request parameters, and then accesses the AI ​​interface of the third-party cloud server. The cloud server retrieves the agent's personality configuration from the role database based on the request parameters and returns it to the device, including: S31. The cloud server obtains and parses the request parameters sent by the device. S32. Determine whether the AI ​​virtual character ID in the request parameters is consistent with the AI ​​virtual character ID currently used on the device. If they are consistent, maintain the current personality configuration and send a role existence notification to the device. Otherwise, proceed to step S33. S33. Search the role database. If the corresponding AI virtual role ID is found, return the personality configuration corresponding to the AI ​​virtual role ID to the device and send a role existence notification to the device. Otherwise, send a no-role-existence notification to the device. In step S4, while the device interacts with the user via voice based on the current personality state parameters, it also displays animations and content corresponding to the personality state and voice.

3. The AI ​​intelligent agent personality switching method based on NFC recognition according to claim 1, characterized in that, Before step S1, the process also includes custom character creation: Configure role parameters on the cloud server and send the role parameters to the devices that interact with the cloud server; The device writes the role parameters into the smart agent's NFC tag in the form of an AI virtual role ID.

4. The AI ​​intelligent agent personality switching method based on NFC recognition according to claim 1, characterized in that, Step S4 is followed by switching the agent's avatar background: Capture images containing environmental scenes and / or clothing, and convert the content of the images into digital information; The device receives the digital information and writes the image material information from the digital information into the image data area of ​​the smart agent's NFC tag; The device re-reads the AI ​​virtual role ID of the intelligent agent and reports the intelligent agent information to the cloud server; The cloud server acquires intelligent agent information, parses and applies new scene data and / or image data from the intelligent agent's image data area; The cloud server switches the scene data and / or appearance data of the intelligent agent, sends the new scene and / or appearance of the role to the device, and the device displays the new scene and / or appearance of the intelligent agent.

5. The AI ​​intelligent agent personality switching method based on NFC recognition according to claim 1, characterized in that, Step S4 is followed by agent role growth updates: The device acquires location information and writes it into the location information area of ​​the smart agent's NFC tag; The device re-reads the AI ​​virtual role ID of the intelligent agent and reports the intelligent agent information to the cloud server; The cloud server acquires and parses the location information content in the location information area of ​​the intelligent agent, determines whether there is any new location information, and updates the agent's location information when there is new location information; The cloud server provides the new location information of the character to the AI ​​character model as setting information. The AI ​​character model learns and updates its growth information based on the new location information to simulate an intelligent agent model that travels and grows with the user. The growth information includes the local customs and culture of the area where the new location is located.

6. The AI ​​intelligent agent personality switching method based on NFC recognition according to claim 1, characterized in that, When the device does not detect an NFC tag signal, or the parsing fails, or the NFC tag recognition is inconsistent, or the number of successful NFC tag recognitions is less than a preset value, it enters a sleep state for a preset duration.

7. The AI ​​intelligent agent personality switching method based on NFC recognition according to claim 1, characterized in that, The device caches the most recent personality configuration when it is offline.

8. An AI intelligent agent personality switching device based on NFC recognition, used to implement the AI ​​intelligent agent personality switching method based on NFC recognition as described in any one of claims 1-6, characterized in that, The AI ​​intelligent agent personality switching device includes a device that interacts with the intelligent agent and a cloud server that interacts with the device. The device includes: The NFC read / write module is used to read the NFC tag signal of the smart agent and to perform write operations to the smart agent to update the NFC tag signal of the smart agent. The information processing module is used to parse the NFC tag signal of the intelligent agent and obtain the AI ​​virtual role ID corresponding to the intelligent agent; The communication module is used to send request parameters containing the AI ​​virtual character ID to the cloud server and receive the personality configuration returned by the cloud server to switch and update the corresponding personality state of the intelligent agent. The audio module is used to acquire and process user voice and play AI voice messages. The cloud server has a character database for storing character personality configurations. The cloud server retrieves the personality configuration of the agent from its character database according to the request parameters and returns it to the communication module.

9. The AI ​​intelligent agent personality switching device based on NFC recognition according to claim 8, characterized in that, The device also includes: The display module is used to display the AI ​​character's appearance, interactive animations, and content. The interaction module provides an interface for interacting with users.

10. The AI ​​intelligent agent personality switching device based on NFC recognition according to claim 8, characterized in that, The communication module interacts with the cloud server through the SDK module.

Images